---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: ground_truth
    dtype: string
  splits:
  - name: train
    num_bytes: 315014535.0
    num_examples: 800
  - name: validation
    num_bytes: 40476324.0
    num_examples: 100
  - name: test
    num_bytes: 38979989.0
    num_examples: 100
  download_size: 394478495
  dataset_size: 394470848.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---